> >  In this manner the release process itself substitutes as the quality
> > assurance.
> 
> In fact, in the early days of Tiles we did like this, but, in fact,
> the "resolved" state is still without meaning.
> Suppose that a build fails, you have to declare the version "invalid"
> and move all the issues to another version. This step is needed even
> if you use "resolved" state.

I don't quite understand you.
In this context what do you mean the "build failed" ? And why would the
issue be invalid?

> >> Not necessarily. If you like, you can deploy a snapshot (mvn deploy),
> >> let the other project depend on the snapshot
> >
> > I'm not very fond of inheriting from SNAPSHOT parents. Apart from the
> > project instability it introduces it also slows down every build.
> 
> Ok for the instability, but it is resolved when it is released thanks
> to the Maven release plugin that checks the presence of snapshots. 

Yes of course. thanks for your patience. 
Then I've made the commit so to use 2-SNAPSHOT, and i'll update the site
latter today.

~mck

-- 
To mess up a Linux box, you need to work at it; to mess up your Windows
box, you just need to work on it. 
| http://semb.wever.org | http://sesat.no | http://finn.no |

Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to